Dissemination of antibiotic-resistant bacteria Advanced methods are needed to identify antibiotic-resistant (AMR) genes in bacterial pathogens. With full-genome sequencing, the best attack methods can be used to identify AMR genes by distinguishing unknown sequences from known AMR sequences in existing online repositories. Researchers at WSU (Washington State University) have developed a new method for identifying previously unknown antibiotic-resistant gene bacteria. Using machine learning and game theory, the researchers were able to determine three different types of antibiotic-resistant genes.
Bacteria can cause bloodstream infections, which can be complicated by increased antibiotic resistance (AMR) in bacterial therapy. AMR increases mortality and increases hospitalization time. Every year millions of people in the United States get infected with AMR bacteria, and thousands die. Therefore, proper identification of bacterial AMR is essential for the proper administration of appropriate antibacterial agents. In vitro cultures are used to monitor the growth of bacteria at different vitamin concentrations for the detection of AMR in bacteria. Also, many bacteria cannot be grown, and metagenomic studies can obtain many of these. Researchers at WSU (Washington State University) have developed a new method for identifying previously unknown antibiotic-resistant gene bacteria.
Researchers including Abu Sayeed Chaudhry, a graduate of the School of Electrical Engineering and Computer Science, and Professor Shira Broschat, and Douglas Cole of the Global Animal Health School at Paul Allen, report in Scientific Reports, a leading journal of their work.
The spread of resistant bacteria is a growing problem worldwide. Every year, millions of people in the United States are infected with drug-resistant pathogens, and thousands die from untreated pneumonia or arterial thinning.
In previous years, researchers have been searching for similar sequences of genes in public databases and working on gene sequencing to identify antibiotic-resistant genes. This operates to identify known antibiotic-resistant genes, but not with unusual genes. “This tool allows us to identify unidentified hypothetical resistance genes based on simple sequential comparisons with public databases.”
The WSU team decided to use the game theory to model strategic interactions among players in several fields, particularly economics, to help identify antibiotic resistance genes.
In-game theory, models determine how one participant’s behavior affects the behavior of the other.
Managing their machine training algorithm & game theory approach, the researchers looked at the communications of some features rather than their sequence similarity, including the physicochemical, evolutionary, & compositional features of genetic material and protein sequences.